RNA-sequence (RNA-Seq) analysis

Confluent AtT20 cells were treated with 1 µM JQ1- or JQ1 for 48 h before RNA extraction using the RNeasy kit (Qiagen). RNA sequencing was performed on three independent experimental replicates for each cell line and treatment. Following polyA-enrichment and library preparation, 50 bp paired-end sequencing was performed (Illumina 2500 machine, rapid mode, 2 lanes). Reads were aligned to the mouse reference genome (GRCm37) using TopHat2 (Kim et al. 2013) and duplicate reads were removed (Picard Tools MarkDuplicates). Approximately 22–30 million high-quality reads per sample were mapped uniquely to Ensembl-annotated genes; gene counts were summarised using HTSeq (Anders et al. 2015) and filtered to exclude genes with fewer than ten reads on average per sample. The filtered set of genes were further analysed in the R statistical software (https://www.r-project.org/) to characterise data quality and sample clustering patterns. Data normalisation and differential expression analysis, comparing JQ1 and JQ1- treatment conditions, were performed using the edgeR package (Robinson et al. 2010). Genes with adjusted P value <0.05 and showing a fold change >2 in either direction were considered significant. Identification of altered cellular pathways was undertaken using QIAGEN Ingenuity® Pathway Analysis (IPA®, QIAGEN, www.qiagen.com/ingenuity). Gene-set enrichment analysis (GSEA) against MSigDB (software.broadinstitute.org/gsea/msigdb) gene-sets (v.6.2) was performed using the fgsea package in R/Bioconductor (http://bioconductor.org/packages/fgsea/) after mapping mouse EntrezIDs to human using the biomaRt package (Durinck et al. 2009, Sergushichev 2016).